[PATCH 2/2] drm/amdgpu: Fix a bug in amdgpu_fill_buffer()

[Yong Zhao]

When max_bytes is not 8 bytes aligned and bo size is larger than
max_bytes, the last 8 bytes in a ttm node may be left unchanged.
For example, on pre SDMA 4.0, max_bytes = 0x1fffff, and the bo size
is 0x200000, the problem will happen.

In order to fix the problem, we store the max nums of PTEs/PDEs
a single operation can set separately in amdgpu_vm_pte_funcs
structure.

Change-Id: I37c588a57cb63f1a8251fb5ead2eff4b39e047c9
Signed-off-by: Yong Zhao <yong.zhao at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     | 6 ++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 8 ++++----
 drivers/gpu/drm/amd/amdgpu/cik_sdma.c   | 3 +++
 drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c  | 3 +++
 drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c  | 3 +++
 drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c  | 3 +++
 drivers/gpu/drm/amd/amdgpu/si_dma.c     | 3 +++
 7 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index a34c4cb..91bb111 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -309,6 +309,12 @@ struct amdgpu_vm_pte_funcs {
 			    uint64_t pe,
 			    uint64_t addr, unsigned count,
 			    uint32_t incr, uint64_t flags);
+
+	/* maximum nums of PTEs/PDEs in a single operation */
+	uint32_t	set_max_nums_pte_pde;
+
+	/* number of dw to reserve per operation */
+	unsigned	set_pte_pde_num_dw;
 };
 
 /* provided by the gmc block */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 2180ed3..8685b0c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1556,8 +1556,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 		       struct dma_fence **fence)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
-	/* max_bytes applies to SDMA_OP_PTEPDE as well as SDMA_OP_CONST_FILL*/
-	uint32_t max_bytes = adev->mman.buffer_funcs->fill_max_bytes;
+	uint32_t max_bytes = 8 *
+			adev->vm_manager.vm_pte_funcs->set_max_nums_pte_pde;
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
 
 	struct drm_mm_node *mm_node;
@@ -1589,8 +1589,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 		++mm_node;
 	}
 
-	/* 10 double words for each SDMA_OP_PTEPDE cmd */
-	num_dw = num_loops * 10;
+	/* num of dwords for each SDMA_OP_PTEPDE cmd */
+	num_dw = num_loops * adev->vm_manager.vm_pte_funcs->set_pte_pde_num_dw;
 
 	/* for IB padding */
 	num_dw += 64;
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index f508f4d..ff59351 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -1390,6 +1390,9 @@ static const struct amdgpu_vm_pte_funcs cik_sdma_vm_pte_funcs = {
 	.copy_pte = cik_sdma_vm_copy_pte,
 	.write_pte = cik_sdma_vm_write_pte,
 	.set_pte_pde = cik_sdma_vm_set_pte_pde,
+
+	.set_max_nums_pte_pde = 0x1fffff >> 3,
+	.set_pte_pde_num_dw = 10,
 };
 
 static void cik_sdma_set_vm_pte_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index f2d0710..aec3586 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -1327,6 +1327,9 @@ static const struct amdgpu_vm_pte_funcs sdma_v2_4_vm_pte_funcs = {
 	.copy_pte = sdma_v2_4_vm_copy_pte,
 	.write_pte = sdma_v2_4_vm_write_pte,
 	.set_pte_pde = sdma_v2_4_vm_set_pte_pde,
+
+	.set_max_nums_pte_pde = 0x1fffff >> 3,
+	.set_pte_pde_num_dw = 10,
 };
 
 static void sdma_v2_4_set_vm_pte_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 4e7fe07..7610272 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -1751,6 +1751,9 @@ static const struct amdgpu_vm_pte_funcs sdma_v3_0_vm_pte_funcs = {
 	.copy_pte = sdma_v3_0_vm_copy_pte,
 	.write_pte = sdma_v3_0_vm_write_pte,
 	.set_pte_pde = sdma_v3_0_vm_set_pte_pde,
+
+	.set_max_nums_pte_pde = 0x3fffe0 >> 3,
+	.set_pte_pde_num_dw = 10,
 };
 
 static void sdma_v3_0_set_vm_pte_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index fd7c72a..6e1e0c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -1717,6 +1717,9 @@ static const struct amdgpu_vm_pte_funcs sdma_v4_0_vm_pte_funcs = {
 	.copy_pte = sdma_v4_0_vm_copy_pte,
 	.write_pte = sdma_v4_0_vm_write_pte,
 	.set_pte_pde = sdma_v4_0_vm_set_pte_pde,
+
+	.set_max_nums_pte_pde = 0x400000 >> 3,
+	.set_pte_pde_num_dw = 10,
 };
 
 static void sdma_v4_0_set_vm_pte_funcs(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index 112969f..2d3091f 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -890,6 +890,9 @@ static const struct amdgpu_vm_pte_funcs si_dma_vm_pte_funcs = {
 	.copy_pte = si_dma_vm_copy_pte,
 	.write_pte = si_dma_vm_write_pte,
 	.set_pte_pde = si_dma_vm_set_pte_pde,
+
+	.set_max_nums_pte_pde = 0xffff8 >> 3,
+	.set_pte_pde_num_dw = 9,
 };
 
 static void si_dma_set_vm_pte_funcs(struct amdgpu_device *adev)
-- 
2.7.4